Chorionic villus sampling and amniocentesis are alternative procedures for detecting fetal chromosomal abnormalities. Given their differing benefits and risks, the "optimal" procedure for pregnant women remains unclear. Our preliminary work, using a decision-analytic model, suggested that the choice of procedure may be "utility-driven", i.e., dependent on the value to the patient of the differing outcomes (e.g., spontaneous abortion, therapeutic abortion, unconfirmed normal and abnormal results, anxiety while awaiting results), rather than on the probabilities of these outcomes. To explore these issues further, we propose to study: 1) the utilities associated with each aspect of the prenatal diagnosis decision, as assessed by pregnant women, and by the physicians who refer them for genetic testing; and 2) the relationship between the choice prescribed by decision-analytic models and the choice actually made by individual women. We will collect data on sociodemographic, obstetric, and health care provider factors for women presenting for genetic counseling prior to amniocentesis or chorionic villus sampling. Patients' and physicians' values for the outcomes of the prenatal diagnosis decision will be assessed using linear rating scales, a standard utility-assessment technique. Using decision-analytic models, with probabilities obtained from the literature that are used during counseling sessions, we will compare the "best" choices based on patients' and physicians' assessed utilities with the choice actually made by each patient. Differences in choices will be examined to determine demographic, obstetric, and patient- and provider-utility covariates that might explain these differences. These studies will yield insight into patients' and physicians' values for the outcomes that follow prenatal diagnostic testing, the contribution of these values to their decision making, and the relationship of their decisions to those prescribed by formal decision-analytic models.